A little History of Paleoclimatology in Europe

Data to reconstruct Late Pleistocene Climate History in Europe originally came from four sources: the study of glacier advances, the analysis of pollen cores, the sedimentology of cave sediments and the sedimentology of loess deposits. Each of this sources has its own limitations:

The advance of a glacier results in a clear readable geomorphological feature (moraine belt), but has, however, the disadvantage of erasing the features left by former advances. Also glaciers tell us little or nothing about intensity and duration of the warm period preceding their last advance.

Pollen bearing sediments are built up best during warm oszillations, during cold stages they very often are disturbed. Reverse is the case for cave sediments: here cold times (producing frost cracking of the cave walls) are best preserved, while warm oszillations lead to a dissolution of the deposited sediments. Best is loess stratigraphy, where cold times produce loess accumulation and warm oszillations lead to soil formation.

From hence one may understand the problems to correlate the results of these different disciplines, each using its own terms, each being confronted with the question of the completeness of their stratigraphies resp. the age of their sections.

Help came from physicists: First to mention the orbital theory detected by Milancovic in the twenties, who computed the duration of three anomalities in the orbit of the earth, which should result in long time changes of the insolation of the earth and thus influence its climate. A second milestone was the invention of radiocarbon dating in the forties by Libby, giving the possibility to check the assumed age of the stratigraphical sections.

A further milestone came in the fifties with the detection of the dependency of oxygene isotope ratio to climate, as observed in shales of micro organisms accumulated in deep sea deposits. Longtime sequences (e.g. half a million of years) displayed rhythms as computed by the orbital theory, which, in reverse, now was used to help in dating. Summarizing the results, SPECMAP gave a first clear vision of the flow of the climatic changes during middle and upper pleistocene.

Deep sea cores, however, suffer from the burrowings of the organisms living in those sediments, so they are not best to resolve short time events. A further milestone came in the early sixties with the detection of isotope changes in athmospherous oxygene, conserved in continetal ice caps (esp. Greenland and Antarctica). In the early nineties two big drilling projects in Greenland where conducted, GRIP and GISP2, each reaching through the whole last glaciation.

GRIP and GISP2 have to be regarded as the breakthrough in paleoclimatic history of Upper Pleistocene and it is of most importance, that there were two enterprises: since they showed the same results, we can be sure of their reliabilty !

But what did they show ? a former unbelievable number of short time oszillations, leading to the designation 'High Frequency Climate Curve'. Continental Paleoclimatologists now could take a relief: They were right with their multitude of oszillations observed. Or, as one core driller stated: "The great number of interstadials, revealed by the ice cores, may be the cause of some of the confusion about the number of interstadials and their timing in northwest European climate records and offer an opportunity for their interpretation" (Grootes et al., Nature 366, 1993, 554).

A little History of this Document

During my whole career there was nothing more frustrating, than seeing those terms of Paleoclimatologists without understanding anything ! Everything seemed contradictory: What is what ? Is it synonymous, or is it only partially synonymous - is it real at all ?

In spring 1996, preparing an article on a surface collection (Vilshofen) with a special type of silex artefacts being characteristic for the Middle Magdalenian, I wanted to have a nice table, showing the assumed chronological position in relation to the paleoclimate-archeologic sequences of Austria/Moravia and South-Western France, both being most important to the understanding of my material, since it comes from a geographical position (SE-Bavaria) lying between these two classic paleolithic landscapes.

So I came about GRIP and GISP2 ice cores, whose data I found ready for a download at NOAA. After a display of the numbers (16/18 ratio, age) by a PASCAL program, I started a survey on paleoclimatological articles, especially on the works of Arl. Leroi-Gourhan (palynologist), H. Laville (cave sedimentologist), B. Bosselin & F. Djindjian (archeologists), B. Klíma (archeologist and loess sedimentologist) and P. Haesaerts (loess sedimentologist). My first attempts to correlate their sequences of climate events to those of the ice core records, however, were really disapointing - until I understood: Oh, they are talking in radiocarbon years !

If the introduction of radiocarbon chronology was a revolution to archaeolgy, the detection, that athmospheric 14C production is not steady, was a revolution to radiocarbon dating itself: Long time calibrations by dendrochronological counting displayed remarkable deviations. Until now dendrochronological calibrating ends at about 10 000 years, since finding trees in late Glacial is difficult. Another way of calibrating therefore was established by using the growth of coral reefs, this calibration going down to 19262 bp, which gives a calibrated age of 22950 years (Bard et al. in Radiocarbon, 35, 1, 1993, 191-199).

It was the estimation of the Laugerie interstadial by Arl. Leroi-Gourhan (19700 to 18500 bp), which showed me, that GISP2 is in accordance with the Bard calibration set: 19700 to 18500 calibrated according Bard results in 23450 to 22100, in GISP2 the Laugerie has an age 23500 to 22100. This result was really encouraging !

My Radiocarbon Scale beyond the Bard Calibration Set

The world of physicists may be pure - real world, however, is dirty ! This phrase one could take as a motto under which my radiocarbon scale was established. Trying to use original radiocarbon dates soon turned out to be dangerous, since a lot of them obviously lay outside the range of the phenomenon to be dated. But it could not be me to decide which one is incorrect. This has to be done by the very specialist for the phenomenon in question. So I just relied on their summarizing tables, built up by their life long experience.

There is one magical date: 43000 bp for the beginning of Hengelo, which I found in the tables of Laville and Bosselin & Djindjian. I am not sure where it comes from (may be from Royat near Puy-de-Dôme, used to date the Tambourets oszillation). By way of trial I correlated this age to the onset of Hengelo in GISP2 (having 45500 GISP2 years) and connected this point with a simple straight line with the end of the Bard calibration set (cf. supra). Proofing this deviation with the estimations of the specialists showed good results and so there was nothing else to do.

Such a simple procedure may cause a distraught state within the reader, so an important remark immediately has to follow: For the moment we cannot be sure, that the time scale of GISP2 data used for Abb. 4 represents consistent calendar years ! It was gained by annual layer counting, but obviously not all of the 40000 layers were counted directly, since also statistic processing (eg. concerning ice pressure) was necessary. Besides this, the problems of the ice core time scales best can be seen with a direct comparison of GISP2 and GRIP (Weißmüller 1997, Abb. 2), which around 30500 diverge with 3700 years !

From this we should deduce: As long as it is incertain, whether the ice core time scales represent reliable calendar years, any attempt to use them for a definite radiocarbon calibration must be elusive. I should also remark, that my original intention was not directed to radiocarbon calibration. My intenion was the correlation of the climatic phenomena, which also can be done with relative time scales. Finally I have to mention, that the long series of new radiocarbon dates (AMS) for the Austria-Moravian soil development, which was published after my correlation was finished, shows remarkable good results.

Recent Radiocarbon Calibration Attempts

Finally its seems useful, to have a look on other attempts of calibrating radiocarbon for the Late Pleistocene. In the table below I have put them all together, so you easily can compare them. The lines made out of points should help you to find those sections, in which the deviation to my scale exceeds 500 years.

BPcal    W.W.      V.A.      K.P.      J.W.      A.V.

10        9.1                 8.8      
11        9.8                10.0
12       10.2                10.4      10.1
13       11.1                11.3      11.0
14       12.0                12.1      11.8
15       12.7                12.5      12.7      12.5 - 13.2
16       13.4                13.5      13.4
17       14.2                14.4      14.2
18       15.1      15.6      15.5      14.9
19       16.1      16.3      16.2      15.7
20       16.9      17.2      17.3      16.5      16.5 - 17.3 
21       17.6      17.9                17.4
22       18.5      18.8      19.0      18.5
23       19.5      19.9                19.6
24       20.6      20.9  ... 20.6 ...  20.6
25       21.7      21.8  :   22.1   :  21.5      21.7 - 22.0
26       22.7      22.8  :   23.3   :. 22.3 ...
27       23.7      23.8  :   24.6      23.0   : 
28       24.7      24.7  :   25.3      23.8   :
29       25.8      25.7  :             24.8   :
30       26.9      26.8  :   26.0      26.2   :  27.0
31       27.9      27.7  :          .. 27.7 ..:
32       29.0      28.7  :          :  29.1   
33       30.0      29.8  :   31.6   :  30.2   ..............
34       31.0      31.0  :   32.9   :  30.9   :
35       32.1      32.2  :   33.2   :. 31.6 ..:  31.4       
36       33.1      33.2  :             31.8
37       34.2      34.4  :   35.1      32.0
38       35.2      35.5  :             32.1
39       36.2      36.7  :             32.3
40       37.2      37.7  :.. 36.2 ..   32.8      32.6 - 34.9
41       38.3  ... 38.8 .... 38.4 .:   33.6
42       39.3  :   39.9                34.6
43       40.3  :   41.2      40.9      35.9
44       41.2  :   42.7      41.8      37.4
45       42.4  :   44.6      42.4 ?    39.1      43.7 - 44.0
               
BPcal    W.W.      V.A.      K.P.      J.W.      A.V.